coordination #102882: [epic] All OSD PPC64LE workers except malbec appear to have horribly broken cache service
Added by okurz almost 3 years ago. Updated almost 3 years ago.
0%
As a first step we should make a list of our switches with firmware versions. Even if they are all Cisco, the IOS commands have changed a bit over time.
I guess https://racktables.nue.suse.com/index.php?andor=and&cft%5B%5D=32&cft%5B%5D=22&cfe=%7B%24typeid_8%7D&page=depot&tab=default could be a complete list. The specific entries would need to be checked. For example qanet01nue.qa shows
HW revision: v02
HW type: Cisco SG300-52
OEM S/N 1: DNI180900QM
SW type: Marvell ROS 1.4.0
SW version: 1.4.7.5
others like qanet12nue.qa show at least a difference in SW type with "Marvell ROS 1.4.1". qanet13nue.qa from our favorite ppc rack shows 1.4.0. Maybe that needs a manual update of the racktables properties.
Effectively I learned something together with and from nicksinger yesterday when being in the SUSE Nbg SRV2 server room connecting machines into the QA racks and debugging in particular a connection problem of the machine "kernel-rt".
I think https://www.cisco.com/c/en/us/td/docs/routers/nfvis/switch_command/b-nfvis-switch-command-reference.html could be a good technical documentation and reference to consult. I added documentation how to reach switches, how to login and the command reference to https://confluence.suse.com/display/openqa/openQA#openQA-AccesstohardwareatSUSE
I asked the team for feedback in https://suse.slack.com/archives/C02AJ1E568M/p1645629237571949 . If there are no objections I will resolve this ticket soon.
I've also put some Cisco courses on my todo list.
No objections, considered done